GLP-1s and mental health: mood, anxiety, and what the research says

Concerns about GLP-1 medications and psychiatric side effects, including suicidal ideation, prompted a formal FDA review. Here's what regulators found, what remains genuinely uncertain, and what to discuss with your doctor if you have a mental health history.

The concern didn't appear out of nowhere. Early case reports and pharmacovigilance signals (systematic monitoring of drug safety data after a medication reaches the market), along with regulatory attention in Europe, raised the possibility that some people taking GLP-1 medications might be experiencing changes in mood or, in some cases, thoughts of self-harm. Patient-reported experiences added to the conversation. Regulators take these kinds of signals seriously even before causation is established. That's how the system is supposed to work: a potential safety concern surfaces, it gets investigated, and the findings guide what happens next. The warning added to GLP-1 labels was a precautionary step while that investigation was underway, not a conclusion. One reason this is genuinely complicated to study: obesity and type 2 diabetes, the conditions these medications most commonly treat, are themselves associated with elevated rates of depression and anxiety. Any attempt to understand whether a GLP-1 medication is affecting someone's mental health has to account for the fact that the underlying condition may already be doing so. Separating those effects is not straightforward.

After completing a formal pharmacovigilance review of GLP-1 receptor agonists, the FDA found no evidence of a causal link to suicidal ideation or behavior. As of February 2026, the suicidal behavior and ideation warning was removed from the prescribing information for both Wegovy and Zepbound. The finding is that a causal link was not detected, not that any possibility of risk has been ruled out with absolute certainty. Pharmacovigilance is an ongoing process, and surveillance continues. The removal of the warning reflects the current state of the evidence, which may be updated as more data become available.

Beyond the question of suicidal ideation, the picture is less settled. Some patients report changes in mood after starting a GLP-1 medication, sometimes positive, sometimes not. Those reports are real and worth taking seriously, but patient reports and observational data can't tell us whether the medication itself is responsible. Here's why that's hard to untangle: when someone starts a GLP-1 medication, a lot of things change at once. They may lose weight. Their energy levels may shift. Their relationship with food changes. Their metabolic health may improve. Any of these factors, independently of the drug itself, could affect how someone feels emotionally. What current evidence has not established:

Whether GLP-1 medications improve or worsen depression or anxiety

What biological mechanisms, if any, might connect these medications to mood

Whether effects differ across subgroups, for example people with pre-existing psychiatric diagnoses or those taking psychiatric medications

Preliminary findings from early or small studies sometimes circulate in the media. The honest answer is that we don't yet know enough to make confident claims about GLP-1s and mental health outcomes beyond what the FDA review addressed.

A 2026 meta-analysis published in BMC Psychiatry pooled data from 19 studies and found that GLP-1 receptor agonist use was associated with a moderately elevated risk of depression (odds ratio 1.49, 95% confidence interval 1.18 to 1.88). The finding was statistically significant, but the analysis also flagged very high heterogeneity across the included studies, meaning the results varied substantially depending on which population was studied, how depression was measured, and how long patients were followed. A pooled number drawn from highly inconsistent data should be interpreted cautiously rather than taken at face value.

A separate 2025 study, also in BMC Psychiatry, approached the question from a different angle entirely. Using Mendelian randomization, a method that uses genetic variants as proxies for drug exposure to test for causal relationships, researchers found no significant genetic associations between GLP-1 receptor agonist exposure and depression. This type of analysis is less vulnerable to the confounding factors that complicate observational research, and its null finding suggests that whatever correlation appears in patient data may not reflect a direct drug effect.

These two lines of evidence point in different directions, and that tension is worth sitting with rather than resolving prematurely. Observational data suggest a possible risk signal. Genetic data suggest no causal link. The honest conclusion is that we cannot yet distinguish a true drug effect from the confounders that surround it, including the mood burden of the conditions GLP-1s treat and the psychological upheaval that accompanies rapid weight change.

There is one thing the science has established clearly: these drugs do act in the brain. Semaglutide is confirmed to be CNS-penetrant, meaning it crosses the blood-brain barrier and accumulates in brain tissue, particularly the area postrema in the dorsal vagal complex, a region involved in nausea and appetite regulation. A 2025 study published in Nature identified a specific population of neurons, called Adcyap1+ neurons, as mediating the weight-loss effects of GLP-1 receptor agonists. None of this proves that central nervous system activity translates into psychiatric effects, but it establishes the biological plausibility that researchers are now working to characterize. The drugs are active in the brain. The open question is what else that central action does.

Not every psychological challenge that accompanies GLP-1 use is a side effect of the medication itself. Some of the most commonly reported difficulties have less to do with pharmacology and more to do with the pace of change.

When the body transforms faster than a person's sense of self can adjust, the result can be a disorienting mismatch between what someone sees in the mirror and how they feel inside. Clinicians have described this informally as body dysmorphia by speed: the physical transformation outpaces the psychological one. Patients report feeling like strangers in their own bodies, struggling to recognize themselves, or feeling uncertain about how to occupy a body that no longer matches their internal map.

A 2025 study published in ScienceDirect examined the relationship between interest in GLP-1 medications and body image. Researchers found that people expressing greater interest in GLP-1 use also reported higher levels of body shame and body surveillance, and lower levels of body appreciation. Notably, body appreciation appeared to serve as a protective factor, meaning that people with a more grounded and accepting relationship to their body before starting treatment seemed better equipped to weather the identity shifts that followed.

These experiences are not captured in adverse event reports and do not appear in prescribing information. They are not drug side effects in the clinical sense. But they are real, they are common, and they are clinically relevant. Psychologists and therapists are increasingly working alongside prescribers to help GLP-1 patients manage identity adjustment alongside physical transformation. For care teams, the takeaway is straightforward: psychological support should be part of the treatment plan, not an afterthought triggered only when something goes visibly wrong.

The intersection of GLP-1 medications and eating disorders is one of the more closely watched areas in current clinical practice.

A 2025 systematic review published in Eating and Weight Disorders found that GLP-1 receptor agonists show promise for binge eating disorder specifically, with studies reporting improved Binge Eating Scale scores and greater weight loss in this population compared to placebo. That is a meaningful finding for a condition with limited pharmacological options.

The concern, however, is what happens beneath the surface. GLP-1 medications reduce physiological appetite, but they do not address the psychological drivers of disordered eating. When binge eating is suppressed pharmacologically, the underlying emotional patterns do not disappear. Clinicians have raised the possibility of symptom substitution: binge eating may be replaced by restriction, compulsive exercise, or other compensatory behaviors that are harder to detect and may go unaddressed if no one is looking for them.

This matters in part because of how common undiagnosed eating disorders are in weight management populations. Estimates suggest that up to one-third of patients seeking weight management treatment may have an eating disorder that has not been formally identified. Without screening, these patients may begin GLP-1 therapy without the individualized support their situation requires.

No current evidence supports the use of GLP-1 receptor agonists as a long-term treatment for eating disorders. This population needs careful clinical attention, and screening for disordered eating before initiating treatment is an important step that care teams should not skip.

Most of the available data on GLP-1s and mental health come from observational studies, research that tracks what happens to people over time but isn't designed to prove cause and effect. Randomized controlled trials specifically designed to measure psychiatric outcomes are limited. Long-term mental health effects have not been established; the medications are relatively new in their current form, and the data don't yet extend far enough to draw conclusions about what happens over years of use. Subgroup data are particularly thin. People with pre-existing psychiatric diagnoses, those taking psychiatric medications, and people with eating disorder histories are the populations where individualized guidance matters most, and where the evidence base is least developed. Research in this area is active, and as larger and longer studies report results, clinical recommendations are likely to become more specific.

If you have a mental health history, a few conversations are worth having before you begin a GLP-1 medication. Share your full psychiatric history, including past episodes of depression, anxiety disorders, any history of suicidal ideation, and any history of eating disorders. Your care team can only make good recommendations if they have the full picture. Ask about your current medications. If you take medication for depression, anxiety, or any other psychiatric condition, ask your prescribing clinician whether there are any interactions to be aware of and how your regimen might need to be monitored. Ask how your team will monitor for mood changes. Starting any new medication warrants a check-in period. Know in advance how often you'll be in contact with your care team and what they'll be watching for. Know who to call if something changes. Before you start, get clear on who to contact if you notice significant mood shifts, changes in sleep, or any thoughts of self-harm. Don't wait for a scheduled appointment if something feels urgent. If you have an eating disorder history, ask specifically. This is an area of clinical concern that warrants a direct conversation. Evidence specific to this population is limited, and individualized guidance matters. Don't stop abruptly without guidance. If you're concerned about how a GLP-1 medication is affecting your mental health, contact your care team rather than stopping on your own. Stopping abruptly should be a shared decision made with medical input.

Once you've started a GLP-1 medication, pay attention to how you're feeling, physically and emotionally. Symptoms worth reporting to your care team promptly include:

Significant changes in mood, especially new or worsening depression

New or worsening anxiety or irritability

Changes in sleep that feel out of the ordinary

Any thoughts of self-harm

Report these symptoms regardless of whether you think they're related to the medication. Your care team can help figure out what's going on, but only if they know. If something feels off, say so.